RTL Code Generation on Syntax-preserving dataset (val)
0.28LossFiFSL
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| FiFSLModel=DeepSeek-7B, Forget sets=10%, Epoch=12025.12 | 0.28 | 55 | 49.6 | 58.3 | |
| Llama-7B BaseModel=Llama-7B2025.12 | 0.3 | 49 | 50.7 | 59.1 | |
| FiFSLModel=Llama-7B, Forget sets=10%, Epoch=12025.12 | 0.3 | 47 | 50.6 | 57.9 | |
| FiFSLModel=Llama-8B, Forget sets=10%, Epoch=12025.12 | 0.31 | 50 | 47 | 55.6 | |
| FiFSLModel=Llama-7B, Forget sets=30%, Epoch=12025.12 | 0.32 | 44 | 49.7 | 56.8 | |
| FiFSLModel=Llama-7B, Forget sets=20%, Epoch=22025.12 | 0.33 | 45 | 48.8 | 56.4 | |
| FiFSLModel=Llama-8B, Forget sets=20%, Epoch=12025.12 | 0.34 | 49 | 44 | 51.8 |